Molecule Correction on ChEBI-20
71.66BLEU ScoreAMREC
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AMRECBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 71.66 | 86.61 | 67.96 | 57.6 | 19.59 | 80.07 | 31.2526 | 7.451 | |
| AMRECBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 70.88 | 85.77 | 63.47 | 54.08 | 16.89 | 79.17 | 20.3243 | 5.1682 | |
| AMRECBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 69.88 | 86.18 | 63.22 | 57.45 | 24.29 | 79.15 | 23.5214 | 9.8556 | |
| ReAct-TBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=true2026.06 | 68.42 | 84.43 | 64.8 | 55.94 | 19.29 | 78.8 | 23.85 | 7.3022 | |
| ReWOO-TBackbone=Claude-haiku-4.5, Tool-augmented=true2026.06 | 66.19 | 83.62 | 58.94 | 49.59 | 14.19 | 75.76 | 22.3514 | 5.6864 | |
| ReWOOBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 65.43 | 84.42 | 61.03 | 53.68 | 23.57 | 75.7 | 27.25 | 8.8451 | |
| PlanAndAct-TBackbone=Claude-haiku-4.5, Tool-augmented=true2026.06 | 64.79 | 80.96 | 59.42 | 48.63 | 9.46 | 75.37 | 22.5 | 5.7312 | |
| ReWOO-TBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=true2026.06 | 64.77 | 84.02 | 59.5 | 53.79 | 22.14 | 75.5 | 27.7143 | 8.7526 | |
| PlanAndAct-TBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=true2026.06 | 64.37 | 80.41 | 60.09 | 50.83 | 12.86 | 76.12 | 25.0929 | 8.3742 | |
| ReAct-TBackbone=Claude-haiku-4.5, Tool-augmented=true2026.06 | 64.28 | 80.88 | 57.58 | 48.14 | 10.14 | 75.44 | 22.5473 | 5.9749 | |
| Initial RawBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 64.07 | — | — | — | — | 72.47 | 46.4433 | — | |
| ReAct-TBackbone=GPT-5.4-mini, Tool-augmented=true2026.06 | 63.83 | 80.79 | 62.76 | 49.15 | 9.79 | 74.89 | 36.7216 | 8.841 | |
| Initial RawBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 63.57 | — | — | — | — | 75.55 | 25.2214 | — | |
| Initial RawBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 63.28 | — | — | — | — | 74.39 | 24.5405 | — | |
| ReWOO-TBackbone=GPT-5.4-mini, Tool-augmented=true2026.06 | 62.14 | 81.52 | 56.72 | 45.09 | 11.34 | 73.83 | 35.8351 | 10.355 | |
| ReWOOBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 62.07 | 82.74 | 58.81 | 47.74 | 14.95 | 74.27 | 36.7887 | 11.3297 | |
| PlanAndAct-TBackbone=GPT-5.4-mini, Tool-augmented=true2026.06 | 61.73 | 80.82 | 60.79 | 47.89 | 7.73 | 73.97 | 39.9021 | 8.4538 | |
| ReWOOBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 61.73 | 80.85 | 53.37 | 45.14 | 13.51 | 74.27 | 23.8243 | 7.2375 | |
| ReActBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 59.02 | 78.86 | 52.15 | 43.46 | 10.14 | 72.02 | 24.8649 | 7.7959 | |
| PlanAndActBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 58.91 | 79.18 | 52.75 | 44.14 | 11.43 | 73.05 | 30.4786 | 10.5412 | |
| ReActBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 58.41 | 79.55 | 51.93 | 44.72 | 17.14 | 71.61 | 29.55 | 10.7484 | |
| LLM-CorrectorBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 56.58 | 79.02 | 49.06 | 41.64 | 12.14 | 70.58 | 34.7214 | 12.9917 | |
| ReActBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 55.74 | 77.66 | 48.73 | 38.23 | 5.15 | 69.54 | 41.7938 | 13.087 | |
| PlanAndActBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 55.16 | 77.7 | 48.82 | 38.02 | 6.19 | 69.45 | 48.3711 | 12.6822 | |
| PlanAndActBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 54.41 | 76.07 | 46.26 | 38.3 | 6.08 | 69.87 | 25.7297 | 9.056 | |
| LLM-CorrectorBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 49.79 | 75.05 | 44.07 | 32.39 | 5.67 | 65.47 | 49.9948 | 14.7324 | |
| SMISELFBackbone=GPT-5.4-mini, Tool-augmented=false2026.06 | 44.67 | 71.54 | 37.08 | 26.28 | 0 | 62.95 | 50.8814 | 18.4483 | |
| SMISELFBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 43.26 | 69.2 | 37.88 | 27.6 | 2.03 | 62.26 | 29.7703 | 13.7593 | |
| SMISELFBackbone=Gemini-3.1-Flash-Lite, Tool-augmented=false2026.06 | 42.47 | 69.32 | 34.93 | 27.29 | 2.14 | 61.72 | 34.1 | 18.3432 | |
| LLM-CorrectorBackbone=Claude-haiku-4.5, Tool-augmented=false2026.06 | 39.13 | 57.04 | 36.62 | 28.97 | 6.76 | 52.13 | 38.3784 | 18.8859 |